A device that performs various monitoring processes for a person in a vehicle by performing image processing on an image shot by an image shooting unit is known. As a processing mode of occupant monitoring that the device has, for example, the determination of the state of an occupant sitting in the driver's seat, the personal authentication of an occupant, and so on are provided. At that time, in order to perform an appropriate monitoring process, an area particularly needed for the monitoring process is desired to be properly captured in the image shot by the image shooting unit. For example, when the processing mode for occupant is the determination of the state of an occupant sitting in the driver's seat, an area where the face of an occupant sitting in the driver's seat exists is the area particularly needed for the monitoring process. When this area is not captured properly, it is difficult to properly determine the state of the occupant sitting in the driver's seat, e.g., drowsy driving, distracted driving, or the like.
For example, in Patent Literature 1, an image processing method including analyzing an acquired image to extract a person's face area, calculating a correction amount that makes the brightness of the extracted face area have an optimum value, and correcting the brightness of the image on the basis of the correction amount is described.